An In Silico Model for Predicting Drug-Induced Hepatotoxicity

Multi-task pretrained language model with novel application domains enables more comprehensive health and ecological toxicity prediction

In silico models for screening substances of healthy and ecological concern are essential for effective chemical management. However, current data-driven toxicity prediction models confront formidable challenges related to expressive capacity, data scarcity, and reliability issues. Thus, this study introduces TOX-BERT, a SMILES-based pretrained model for screening health and ecological toxicity. Results show that masked atom recovery pretraining and multi-task learning offer promising solutions to enhance model capacity and address data scarcity issues. Two novel application domain (AD) parameters, termed PCA-AD and LDS, were proposed to improve prediction reliability of TOX-BERT with accuracy surpassing 90 % and mean absolute error (MAE) below 0.52. TOX-BERT was applied to 18,905 IECSC chemicals, revealing distinct toxicity relationships that align with experimental studies such as those between cardiotoxicity and acute ecotoxicity. In addition to previous PBT screening, 156 potential high-risk chemicals for specific endpoint were identified covering 7 categories. Furthermore, a SMILES-based toxicity site detection approach was developed for structural toxicity analysis. These advancements carry profound implications to address challenges faced by current data-driven toxicity prediction models. TOX-BERT emerges as a valuable tool for more comprehensive, reliable, and applicable predictions of health and ecological toxicity in chemical risk assessment and management.

A computational framework to in silico screen for drug-induced hepatocellular toxicity

Drug-induced liver injury (DILI) is the most common trigger for acute liver failure and the leading cause of attrition in drug development. In this study, we developed an in silico framework to screen drug-induced hepatocellular toxicity (INSIGHT) by integrating the post-treatment transcriptomic data from both rodent models and primary human hepatocytes. We first built an early prediction model using logistic regression with elastic net regularization for 123 compounds and established the INSIGHT framework that can screen for drug-induced hepatotoxicity. The 235 signature genes identified by INSIGHT were involved in metabolism, bile acid synthesis, and stress response pathways. Applying the INSIGHT to an independent transcriptomic dataset treated by 185 compounds predicted that 27 compounds show a high DILI risk, including zoxazolamine and emetine. Further integration with cell image data revealed that predicted compounds with high DILI risk can induce abnormal morphological changes in the endoplasmic reticulum and mitochondrion. Clustering analysis of the treatment-induced transcriptomic changes delineated distinct DILI mechanisms induced by these compounds. Our study presents a computational framework for a mechanistic understanding of long-term liver injury and the prospective prediction of DILI risk.

Improved Detection of Drug-Induced Liver Injury by Integrating Predicted In Vivo and In Vitro Data

Drug-induced liver injury (DILI) has been a significant challenge in drug discovery, often leading to clinical trial failures and necessitating drug withdrawals. Over the last decade, the existing suite of in vitro proxy-DILI assays has generally improved at identifying compounds with hepatotoxicity. However, there is considerable interest in enhancing the in silico prediction of DILI because it allows for evaluating large sets of compounds more quickly and cost-effectively, particularly in the early stages of projects. In this study, we aim to study ML models for DILI prediction that first predict nine proxy-DILI labels and then use them as features in addition to chemical structural features to predict DILI. The features include in vitro (e.g., mitochondrial toxicity, bile salt export pump inhibition) data, in vivo (e.g., preclinical rat hepatotoxicity studies) data, pharmacokinetic parameters of maximum concentration, structural fingerprints, and physicochemical parameters. We trained DILI-prediction models on 888 compounds from the DILI data set (composed of DILIst and DILIrank) and tested them on a held-out external test set of 223 compounds from the DILI data set. The best model, DILIPredictor, attained an AUC-PR of 0.79. This model enabled the detection of the top 25 toxic compounds (2.68 LR+, positive likelihood ratio) compared to models using only structural features (1.65 LR+ score). Using feature interpretation from DILIPredictor, we identified the chemical substructures causing DILI and differentiated cases of DILI caused by compounds in animals but not in humans. For example, DILIPredictor correctly recognized 2-butoxyethanol as nontoxic in humans despite its hepatotoxicity in mice models. Overall, the DILIPredictor model improves the detection of compounds causing DILI with an improved differentiation between animal and human sensitivity and the potential for mechanism evaluation. DILIPredictor required only chemical structures as input for prediction and is publicly available at https://broad.io/DILIPredictor for use via web interface and with all code available for download.

Improved Detection of Drug-Induced Liver Injury by Integrating Predicted in vivo and in vitro Data

Drug-induced liver injury (DILI) has been significant challenge in drug discovery, often leading to clinical trial failures and necessitating drug withdrawals. The existing suite of in vitro proxy-DILI assays is generally effective at identifying compounds with hepatotoxicity. However, there is considerable interest in enhancing in silico prediction of DILI because it allows for the evaluation of large sets of compounds more quickly and cost-effectively, particularly in the early stages of projects. In this study, we aim to study ML models for DILI prediction that first predicts nine proxy-DILI labels and then uses them as features in addition to chemical structural features to predict DILI. The features include in vitro (e.g., mitochondrial toxicity, bile salt export pump inhibition) data, in vivo (e.g., preclinical rat hepatotoxicity studies) data, pharmacokinetic parameters of maximum concentration, structural fingerprints, and physicochemical parameters. We trained DILI-prediction models on 888 compounds from the DILIst dataset and tested on a held-out external test set of 223 compounds from DILIst dataset. The best model, DILIPredictor, attained an AUC-ROC of 0.79. This model enabled the detection of top 25 toxic compounds compared to models using only structural features (2.68 LR+ score). Using feature interpretation from DILIPredictor, we were able to identify the chemical substructures causing DILI as well as differentiate cases DILI is caused by compounds in animals but not in humans. For example, DILIPredictor correctly recognized 2-butoxyethanol as non-toxic in humans despite its hepatotoxicity in mice models. Overall, the DILIPredictor model improves the detection of compounds causing DILI with an improved differentiation between animal and human sensitivity as well as the potential for mechanism evaluation. DILIPredictor is publicly available at https://broad.io/DILIPredictor for use via web interface and with all code available for download and local implementation via https://pypi.org/project/dilipred/.

Enhancing property and activity prediction and interpretation using multiple molecular graph representations with MMGX

Graph Neural Networks (GNNs) excel in compound property and activity prediction, but the choice of molecular graph representations significantly influences model learning and interpretation. While atom-level molecular graphs resemble natural topology, they overlook key substructures or functional groups and their interpretation partially aligns with chemical intuition. Recent research suggests alternative representations using reduced molecular graphs to integrate higher-level chemical information and leverages both representations for model. However, there is a lack of studies about applicability and impact of different molecular graphs on model learning and interpretation. Here, we introduce MMGX (Multiple Molecular Graph eXplainable discovery), investigating the effects of multiple molecular graphs, including Atom, Pharmacophore, JunctionTree, and FunctionalGroup, on model learning and interpretation with various perspectives. Our findings indicate that multiple graphs relatively improve model performance, but in varying degrees depending on datasets. Interpretation from multiple graphs in different views provides more comprehensive features and potential substructures consistent with background knowledge. These results help to understand model decisions and offer valuable insights for subsequent tasks. The concept of multiple molecular graph representations and diverse interpretation perspectives has broad applicability across tasks, architectures, and explanation techniques, enhancing model learning and interpretation for relevant applications in drug discovery.

In vitro-in silico pharmacology and chemistry of Stercularin, isolated from Sterculia diversifolia

Stercularin is a coumarin, isolated from the ethyl acetate fraction of stem bark and leaves of S. diversifolia. Pharmacologically it is active against cancer, diabetes, and inflammation etc. The molecule is further screened for in vitro pharmacological activities. In addition, a detailed description on its drug likeness and pharmacokinetic profile has been established to further explore its fate as a drug candidate. Stercularin exhibited antiglycation, immunomodulatory, and leishmanicidal activity in three different in vitro models. The IC50 values obtained in these three assays were 80.22 ± 0.46 mg/ml, 12.8 ± 1.6 μg/ml, and 8.32 ± 0.42 μg/ml, respectively. In case of drug likeness evaluation, Stercularin has acceptable physicochemical properties and compliant with major drug likeness descriptors i.e., Lipinski rule, Pfizer rule, GSK rule, and "golden triangle". Accepting Lipinski rule implies the oral drug development of Stercularin. Pharmacokinetically, Stercularin is permeable to Caco-2 and MDCK cell lines. 'Boiled-egg' plot suggest intestinal route of absorption, blood brain barrier nonpermeating, and not affected by p-glycoprotein. Stercularin has high plasma protein binding with low free fraction circulating in the plasma. Stercularin proved to be the substrate and/or inhibitor of CYP 450 system with a moderate half-life and clearance rate to allow flexible dosing regimen. Finally, slight risk of toxicity exists for Stercularin, but not being limiting factors of drug knock out. A nature isolated Stercularin possess pharmacological activities and is predicted to have acceptable pharmacokinetic profile. Further drug development and in vivo studies are desirable for optimization.

A Comparison of Machine Learning Approaches for predicting Hepatotoxicity potential using Chemical Structure and Targeted Transcriptomic Data

Animal toxicity testing is time and resource intensive, making it difficult to keep pace with the number of substances requiring assessment. Machine learning (ML) models that use chemical structure information and high-throughput experimental data can be helpful in predicting potential toxicity . However, much of the toxicity data used to train ML models is biased with an unequal balance of positives and negatives primarily since substances selected for in vivo testing are expected to elicit some toxicity effect. To investigate the impact this bias had on predictive performance, various sampling approaches were used to balance in vivo toxicity data as part of a supervised ML workflow to predict hepatotoxicity outcomes from chemical structure and/or targeted transcriptomic data. From the chronic, subchronic, developmental, multigenerational reproductive, and subacute repeat-dose testing toxicity outcomes with a minimum of 50 positive and 50 negative substances, 18 different study-toxicity outcome combinations were evaluated in up to 7 ML models. These included Artificial Neural Networks, Random Forests, Bernouilli Naïve Bayes, Gradient Boosting, and Support Vector classification algorithms which were compared with a local approach, Generalised Read-Across (GenRA), a similarity-weighted k-Nearest Neighbour (k-NN) method. The mean CV F1 performance for unbalanced data across all classifiers and descriptors for chronic liver effects was 0.735 (0.0395 SD). Mean CV F1 performance dropped to 0.639 (0.073 SD) with over-sampling approaches though the poorer performance of KNN approaches in some cases contributed to the observed decrease (mean CV F1 performance excluding KNN was 0.697 (0.072 SD)). With under-sampling approaches, the mean CV F1 was 0.523 (0.083 SD). For developmental liver effects, the mean CV F1 performance was much lower with 0.089 (0.111 SD) for unbalanced approaches and 0.149 (0.084 SD) for under-sampling. Over-sampling approaches led to an increase in mean CV F1 performance (0.234, (0.107 SD)) for developmental liver toxicity. Model performance was found to be dependent on dataset, model type, balancing approach and feature selection. Accordingly tailoring ML workflows for predicting toxicity should consider class imbalance and rely on simpler classifiers first.

Deep Learning Algorithm Based on Molecular Fingerprint for Prediction of Drug-Induced Liver Injury

Drug-induced liver injury (DILI) is one the rare adverse drug reaction (ADR) and multifactorial endpoints. Current preclinical animal models struggle to anticipate it, and in silico methods have emerged as a way with significant potential for doing so. In this study, a high-quality dataset of 1573 compounds was assembled. The 48 classification models, which depended on six different molecular fingerprints, were built via deep neural network (DNN) and seven machine learning algorithms. Comparing the results of the DNN and machine learning models, the optional performing model was found as the one developed based on the DNN with ECFP_6 as input, which achieved the area under the receiver operating characteristic curve (AUC) of 0.713, balanced accuracy (BA) of 0.680, and F1 of 0.753. In addition, we used the SHapley Additive exPlanations (SHAP) algorithm to interpret the models, identified the crucial structural fragments related to DILI risk, and selected the top ten substructures with the highest contribution rankings to serve as warning indicators for subsequent drug hepatotoxicity screening studies. The study demonstrates that the DNN models developed based on molecular fingerprints can be a trustworthy and efficient tool for determining the risk of DILI during the pre-development of novel medications.

GeoDILI: A Robust and Interpretable Model for Drug-Induced Liver Injury Prediction Using Graph Neural Network-Based Molecular Geometric Representation

Drug-induced liver injury (DILI) is a significant cause of drug failure and withdrawal due to liver damage. Accurate prediction of hepatotoxic compounds is crucial for safe drug development. Several DILI prediction models have been published, but they are built on different data sets, making it difficult to compare model performance. Moreover, most existing models are based on molecular fingerprints or descriptors, neglecting molecular geometric properties and lacking interpretability. To address these limitations, we developed GeoDILI, an interpretable graph neural network that uses a molecular geometric representation. First, we utilized a geometry-based pretrained molecular representation and optimized it on the DILI data set to improve predictive performance. Second, we leveraged gradient information to obtain high-precision atomic-level weights and deduce the dominant substructure. We benchmarked GeoDILI against recently published DILI prediction models, as well as popular GNN models and fingerprint-based machine learning models using the same data set, showing superior predictive performance of our proposed model. We applied the interpretable method in the DILI data set and derived seven precise and mechanistically elucidated structural alerts. Overall, GeoDILI provides a promising approach for accurate and interpretable DILI prediction with potential applications in drug discovery and safety assessment. The data and source code are available at GitHub repository (https://github.com/CSU-QJY/GeoDILI).

Usage of model combination in computational toxicology

New Approach Methodologies (NAMs) have ushered in a new era in the field of toxicology, aiming to replace animal testing. However, despite these advancements, they are not exempt from the inherent complexities associated with the study's endpoint. In this review, we have identified three major groups of complexities: mechanistic, chemical space, and methodological. The mechanistic complexity arises from interconnected biological processes within a network that are challenging to model in a single step. In the second group, chemical space complexity exhibits significant dissimilarity between compounds in the training and test series. The third group encompasses algorithmic and molecular descriptor limitations and typical class imbalance problems. To address these complexities, this work provides a guide to the usage of a combination of predictive Quantitative Structure-Activity Relationship (QSAR) models, known as metamodels. This combination of low-level models (LLMs) enables a more precise approach to the problem by focusing on different sub-mechanisms or sub-processes. For mechanistic complexity, multiple Molecular Initiating Events (MIEs) or levels of information are combined to form a mechanistic-based metamodel. Regarding the complexity arising from chemical space, two types of approaches were reviewed to construct a fragment-based chemical space metamodel: those with and without structure sharing. Metamodels with structure sharing utilize unsupervised strategies to identify data patterns and build low-level models for each cluster, which are then combined. For situations without structure sharing due to pharmaceutical industry intellectual property, the use of prediction sharing, and federated learning approaches have been reviewed. Lastly, to tackle methodological complexity, various algorithms are combined to overcome their limitations, diverse descriptors are employed to enhance problem definition and balanced dataset combinations are used to address class imbalance issues (methodological-based metamodels). Remarkably, metamodels consistently outperformed classical QSAR models across all cases, highlighting the importance of alternatives to classical QSAR models when faced with such complexities.

Review of machine learning and deep learning models for toxicity prediction

The ever-increasing number of chemicals has raised public concerns due to their adverse effects on human health and the environment. To protect public health and the environment, it is critical to assess the toxicity of these chemicals. Traditional in vitro and in vivo toxicity assays are complicated, costly, and time-consuming and may face ethical issues. These constraints raise the need for alternative methods for assessing the toxicity of chemicals. Recently, due to the advancement of machine learning algorithms and the increase in computational power, many toxicity prediction models have been developed using various machine learning and deep learning algorithms such as support vector machine, random forest, k-nearest neighbors, ensemble learning, and deep neural network. This review summarizes the machine learning- and deep learning-based toxicity prediction models developed in recent years. Support vector machine and random forest are the most popular machine learning algorithms, and hepatotoxicity, cardiotoxicity, and carcinogenicity are the frequently modeled toxicity endpoints in predictive toxicology. It is known that datasets impact model performance. The quality of datasets used in the development of toxicity prediction models using machine learning and deep learning is vital to the performance of the developed models. The different toxicity assignments for the same chemicals among different datasets of the same type of toxicity have been observed, indicating benchmarking datasets is needed for developing reliable toxicity prediction models using machine learning and deep learning algorithms. This review provides insights into current machine learning models in predictive toxicology, which are expected to promote the development and application of toxicity prediction models in the future.

Pharmacokinetic correlation of structurally modified chalcone derivatives as promising leads to treat tuberculosis

In this study, we evaluated the potential of curated structurally modified chalcone derivatives as anti-tuberculosis (TB) agents through computer-aided drug design. Compounds from the flavonoid family known as chalcones were identified by the chemical group 1,3-diaryl-2-propen-1-one. After a search of the literature, 14 outstanding structurally modified chalcones were selected and evaluated for inhibitory activity against Mycobacterium tuberculosis H37Rv targets. The therapeutic potential of the chalcones was directly based on the drug-likeness and pharmacokinetic properties of the synthesized compounds. Prompt drug selection and personalized therapy are required to prevent TB from progressing and spreading to others. Pharmacokinetic parameters helps in the identification of lead molecule, at the earlier stages of drug development.

Collaborative SAR Modeling and Prospective In Vitro Validation of Oxidative Stress Activation in Human HepG2 Cells

Oxidative stress is the consequence of an abnormal increase of reactive oxygen species (ROS). ROS are generated mainly during the metabolism in both normal and pathological conditions as well as from exposure to xenobiotics. Xenobiotics can, on the one hand, disrupt molecular machinery involved in redox processes and, on the other hand, reduce the effectiveness of the antioxidant activity. Such dysregulation may lead to oxidative damage when combined with oxidative stress overpassing the cell capacity to detoxify ROS. In this work, a green fluorescent protein (GFP)-tagged nuclear factor erythroid 2-related factor 2 (NRF2)-regulated sulfiredoxin reporter (Srxn1-GFP) was used to measure the antioxidant response of HepG2 cells to a large series of drug and drug-like compounds (2230 compounds). These compounds were then classified as positive or negative depending on cellular response and distributed among different modeling groups to establish structure-activity relationship (SAR) models. A selection of models was used to prospectively predict oxidative stress induced by a new set of compounds subsequently experimentally tested to validate the model predictions. Altogether, this exercise exemplifies the different challenges of developing SAR models of a phenotypic cellular readout, model combination, chemical space selection, and results interpretation.

In silico modeling-based new alternative methods to predict drug and herb-induced liver injury: A review

New approach methods (NAMs) have been developed to predict a wide range of toxicities through innovative technologies. Liver injury is one of the most extensively studied endpoints due to its severity and frequency, occurring among populations that consume drugs or dietary supplements. In this review, we focus on recent developments of in silico modeling for liver injury prediction using deep learning and in vitro data based on adverse outcome pathways (AOPs). Despite these models being mainly developed using datasets generated from drug-like molecules, they were also applied to the prediction of hepatotoxicity caused by herbal products. As deep learning has achieved great success in many different fields, advanced machine learning algorithms have been actively applied to improve the accuracy of in silico models. Additionally, the development of liver AOPs, combined with big data in toxicology, has been valuable in developing in silico models with enhanced predictive performance and interpretability. Specifically, one approach involves developing structure-based models for predicting molecular initiating events of liver AOPs, while others use in vitro data with structure information as model inputs for making predictions. Even though liver injury remains a difficult endpoint to predict, advancements in machine learning algorithms and the expansion of in vitro databases with relevant biological knowledge have made a huge impact on improving in silico modeling for drug-induced liver injury prediction.

Machine Learning-Based Prediction of Drug-Induced Hepatotoxicity: An OvA-QSTR Approach

Drug-induced hepatotoxicity, also known as drug-induced liver injury (DILI), is among the possible adverse effects of pharmacotherapy. This clinical condition is accepted as one of the factors leading to patient mortality and morbidity. The LiverTox database was built by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) to predict potential liver damage from medications and take appropriate precautions. The database has classified medicines into seven risk categories (A, B, C, D, E, E*, and X) to avoid medicine-induced liver toxicity. The hepatic damage risk decreases from group A to group E. This study did not include the E* and X classes because they contained unverified and unknown data groups. Our study aims to predict potential liver damage of new drug molecules without using experimental animals. We predict which of the LiverTox risk category drugs with unknown liver toxicity potential will fall into using our one-vs-all quantitative structure-toxicity relationship (OvA-QSTR) model. Our dataset, consisting of 678 organic drug molecules from different pharmacological classes, was collected from LiverTox. The OvA-QSTR models implemented by Bayesian Network (BayesNet) performed well based on the selected descriptors, with the precision-recall curve (PRC) areas ranging from 0.718 to 0.869. Our OvA-QSTR models provide a reliable premarketing risk evaluation of pharmaceutical-induced liver damage potential and offer predictions for different risk levels in DILI.

AI/ML Models to Predict the Severity of Drug-Induced Liver Injury for Small Molecules

Drug-induced liver injury (DILI), believed to be a multifactorial toxicity, has been a leading cause of attrition of small molecules during discovery, clinical development, and postmarketing. Identification of DILI risk early reduces the costs and cycle times associated with drug development. In recent years, several groups have reported predictive models that use physicochemical properties or in vitro and in vivo assay endpoints; however, these approaches have not accounted for liver-expressed proteins and drug molecules. To address this gap, we have developed an integrated artificial intelligence/machine learning (AI/ML) model to predict DILI severity for small molecules using a combination of physicochemical properties and off-target interactions predicted in silico. We compiled a data set of 603 diverse compounds from public databases. Among them, 164 were categorized as Most DILI (M-DILI), 245 as Less DILI (L-DILI), and 194 as No DILI (N-DILI) by the FDA. Six machine learning methods were used to create a consensus model for predicting the DILI potential. These methods include k-nearest neighbor (k-NN), support vector machine (SVM), random forest (RF), Naïve Bayes (NB), artificial neural network (ANN), logistic regression (LR), weighted average ensemble learning (WA) and penalized logistic regression (PLR). Among the analyzed ML methods, SVM, RF, LR, WA, and PLR identified M-DILI and N-DILI compounds, achieving a receiver operating characteristic area under the curve of 0.88, sensitivity of 0.73, and specificity of 0.9. Approximately 43 off-targets, along with physicochemical properties (fsp³, log S, basicity, reactive functional groups, and predicted metabolites), were identified as significant factors in distinguishing between M-DILI and N-DILI compounds. The key off-targets that we identified include: PTGS1, PTGS2, SLC22A12, PPARγ, RXRA, CYP2C9, AKR1C3, MGLL, RET, AR, and ABCC4. The present AI/ML computational approach therefore demonstrates that the integration of physicochemical properties and predicted on- and off-target biological interactions can significantly improve DILI predictivity compared to chemical properties alone.

The System of Self-Consistent Models: QSAR Analysis of Drug-Induced Liver Toxicity

Removing a drug-like substance that can cause drug-induced liver injury from the drug discovery process is a significant task for medicinal chemistry. In silico models can facilitate this process. Semi-correlation is an approach to building in silico models representing the prediction in the active (1)-inactive (0) format. The so-called system of self-consistent models has been suggested as an approach for two tasks: (i) building up a model and (ii) estimating its predictive potential. However, this approach has been tested so far for regression models. Here, the approach is applied to building up and estimating a categorical hepatotoxicity model using the CORAL software. This new process yields good results: sensitivity = 0.77, specificity = 0.75, accuracy = 0.76, and Matthew correlation coefficient = 0.51 (all compounds) and sensitivity = 0.83, specificity = 0.81, accuracy = 0.83 and Matthew correlation coefficient = 0.63 (validation set).

Artificial Intelligence in Drug Toxicity Prediction: Recent Advances, Challenges, and Future Perspectives

Toxicity prediction is a critical step in the drug discovery process that helps identify and prioritize compounds with the greatest potential for safe and effective use in humans, while also reducing the risk of costly late-stage failures. It is estimated that over 30% of drug candidates are discarded owing to toxicity. Recently, artificial intelligence (AI) has been used to improve drug toxicity prediction as it provides more accurate and efficient methods for identifying the potentially toxic effects of new compounds before they are tested in human clinical trials, thus saving time and money. In this review, we present an overview of recent advances in AI-based drug toxicity prediction, including the use of various machine learning algorithms and deep learning architectures, of six major toxicity properties and Tox21 assay end points. Additionally, we provide a list of public data sources and useful toxicity prediction tools for the research community and highlight the challenges that must be addressed to enhance model performance. Finally, we discuss future perspectives for AI-based drug toxicity prediction. This review can aid researchers in understanding toxicity prediction and pave the way for new methods of drug discovery.

Physicochemical Properties, Drug Likeness, ADMET, DFT Studies and in vitro antioxidant activity of Oxindole Derivatives

Poor pharmacokinetic and safety profiles create significant hurdles in the drug development process. This work focuses on a detailed understanding of drug discovery interplay among physicochemical, pharmacokinetic, toxicity endpoints, and antioxidant properties of oxindole derivatives. DFT compıutations were also performed at B3LYP/6-311G** level to evaluate the physicochemical properties, global reactivity features, and intramolecular interactions. The BOILED-Egg pharmacokinetic model envisaged gastrointestinal absorption, blood-brain barrier penetration, and no interaction with p-glycoprotein for compounds C1 and C2. The physicochemical evaluation revealed that C1 possesses superior drug-like properties fit for oral absorption. Both derivatives were predicted to have high plasma protein binding, efficient distribution, and inhibiting CYP 450 major isoforms but serve as substrates only for a few of them. Both molecules have mild to moderate clearance rates. Out of ten toxicity parameters, only hepatotoxicity was predicted. DFT results implied that the meta position of the -OH group made the possibility of charge transfer greater than -para positioned -OH, due to the ΔN_max (eV) values of molecules C1 and C2 being calculated at 2.596 and 2.477, respectively. Both C1 and C2 exhibited a concentration dependant DPPH and ABTS radical scavenging activity. The chemical structure-physicochemical-pharmacokinetic relationship identified the meta position as the favorite for the electron-withdrawing hydroxyl group. This provides useful insight to medicinal chemists to design 6-chlorooxindole derivatives with an acceptable drug-like and pharmacokinetic property.

Towards Decoding Hepatotoxicity of Approved Drugs through Navigation of Multiverse and Consensus Chemical Spaces

Drug-induced liver injury (DILI) is the principal reason for failure in developing drug candidates. It is the most common reason to withdraw from the market after a drug has been approved for clinical use. In this context, data from animal models, liver function tests, and chemical properties could complement each other to understand DILI events better and prevent them. Since the chemical space concept improves decision-making drug design related to the prediction of structure-property relationships, side effects, and polypharmacology drug activity (uniquely mentioning the most recent advances), it is an attractive approach to combining different phenomena influencing DILI events (e.g., individual "chemical spaces") and exploring all events simultaneously in an integrated analysis of the DILI-relevant chemical space. However, currently, no systematic methods allow the fusion of a collection of different chemical spaces to collect different types of data on a unique chemical space representation, namely "consensus chemical space." This study is the first report that implements data fusion to consider different criteria simultaneously to facilitate the analysis of DILI-related events. In particular, the study highlights the importance of analyzing together in vitro and chemical data (e.g., topology, bond order, atom types, presence of rings, ring sizes, and aromaticity of compounds encoded on RDKit fingerprints). These properties could be aimed at improving the understanding of DILI events.

Computational pharmacology and computational chemistry of 4-hydroxyisoleucine: Physicochemical, pharmacokinetic, and DFT-based approaches

Computational pharmacology and chemistry of drug-like properties along with pharmacokinetic studies have made it more amenable to decide or predict a potential drug candidate. 4-Hydroxyisoleucine is a pharmacologically active natural product with prominent antidiabetic properties. In this study, ADMETLab 2.0 was used to determine its important drug-related properties. 4-Hydroxyisoleucine is compliant with important drug-like physicochemical properties and pharma giants' drug-ability rules like Lipinski's, Pfizer, and GlaxoSmithKline (GSK) rules. Pharmacokinetically, it has been predicted to have satisfactory cell permeability. Blood-brain barrier permeation may add central nervous system (CNS) effects, while a very slight probability of being CYP2C9 substrate exists. None of the well-known toxicities were predicted in silico, being congruent with wet lab results, except for a "very slight risk" for respiratory toxicity predicted. The molecule is non ecotoxic as analyzed with common indicators such as bioconcentration and LC₅₀ for fathead minnow and daphnia magna. The toxicity parameters identified 4-hydroxyisoleucine as non-toxic to androgen receptors, PPAR-γ, mitochondrial membrane receptor, heat shock element, and p53. However, out of seven parameters, not even a single toxicophore was found. The density functional theory (DFT) study provided support to the findings obtained from drug-like property predictions. Hence, it is a very logical approach to proceed further with a detailed pharmacokinetics and drug development process for 4-hydroxyisoleucine.

TOXRIC: a comprehensive database of toxicological data and benchmarks

The toxic effects of compounds on environment, humans, and other organisms have been a major focus of many research areas, including drug discovery and ecological research. Identifying the potential toxicity in the early stage of compound/drug discovery is critical. The rapid development of computational methods for evaluating various toxicity categories has increased the need for comprehensive and system-level collection of toxicological data, associated attributes, and benchmarks. To contribute toward this goal, we proposed TOXRIC (https://toxric.bioinforai.tech/), a database with comprehensive toxicological data, standardized attribute data, practical benchmarks, informative visualization of molecular representations, and an intuitive function interface. The data stored in TOXRIC contains 113 372 compounds, 13 toxicity categories, 1474 toxicity endpoints covering in vivo/in vitro endpoints and 39 feature types, covering structural, target, transcriptome, metabolic data, and other descriptors. All the curated datasets of endpoints and features can be retrieved, downloaded and directly used as output or input to Machine Learning (ML)-based prediction models. In addition to serving as a data repository, TOXRIC also provides visualization of benchmarks and molecular representations for all endpoint datasets. Based on these results, researchers can better understand and select optimal feature types, molecular representations, and baseline algorithms for each endpoint prediction task. We believe that the rich information on compound toxicology, ML-ready datasets, benchmarks and molecular representation distribution can greatly facilitate toxicological investigations, interpretation of toxicological mechanisms, compound/drug discovery and the development of computational methods.

Prediction of drug-induced liver injury and cardiotoxicity using chemical structure and in vitro assay data

Drug-induced liver injury (DILI) and cardiotoxicity (DICT) are major adverse effects triggered by many clinically important drugs. To provide an alternative to in vivo toxicity testing, the U.S. Tox21 consortium has screened a collection of ∼10K compounds, including drugs in clinical use, against >70 cell-based assays in a quantitative high-throughput screening (qHTS) format. In this study, we compiled reference compound lists for DILI and DICT and compared the potential of Tox21 assay data with chemical structure information in building prediction models for human in vivo hepatotoxicity and cardiotoxicity. Models were built with four different machine learning algorithms (e.g., Random Forest, Naïve Bayes, eXtreme Gradient Boosting, and Support Vector Machine) and model performance was evaluated by calculating the area under the receiver operating characteristic curve (AUC-ROC). Chemical structure-based models showed reasonable predictive power for DILI (best AUC-ROC = 0.75 ± 0.03) and DICT (best AUC-ROC = 0.83 ± 0.03), while Tox21 assay data alone only showed better than random performance. DILI and DICT prediction models built using a combination of assay data and chemical structure information did not have a positive impact on model performance. The suboptimal predictive performance of the assay data is likely due to insufficient coverage of an adequately predictive number of toxicity mechanisms. The Tox21 consortium is currently expanding coverage of biological response space with additional assays that probe toxicologically important targets and under-represented pathways that may improve the prediction of in vivo toxicity such as DILI and DICT.

Identification of hepatoprotective traditional Chinese medicines based on the structure–activity relationship, molecular network, and machine learning techniques

The efforts focused on discovering potential hepatoprotective drugs are critical for relieving the burdens caused by liver diseases. Traditional Chinese medicine (TCM) is an important resource for discovering hepatoprotective agents. Currently, there are hundreds of hepatoprotective products derived from TCM available in the literature, providing crucial clues to discover novel potential hepatoprotectants from TCMs based on predictive research. In the current study, a large-scale dataset focused on TCM-induced hepatoprotection was established, including 676 hepatoprotective ingredients and 205 hepatoprotective TCMs. Then, a comprehensive analysis based on the structure–activity relationship, molecular network, and machine learning techniques was performed at molecular and holistic TCM levels, respectively. As a result, we developed an in silico model for predicting the hepatoprotective activity of ingredients derived from TCMs, in which the accuracy exceeded 85%. In addition, we originally proposed a material basis and a drug property-based approach to identify potential hepatoprotective TCMs. Consequently, a total of 12 TCMs were predicted to hold potential hepatoprotective activity, nine of which have been proven to be beneficial to the liver in previous publications. The high rate of consistency between our predictive results and the literature reports demonstrated that our methods were technically sound and reliable. In summary, systematical predictive research focused on the hepatoprotection of TCM was conducted in this work, which would not only assist screening of potential hepatoprotectants from TCMs but also provide a novel research mode for discovering the potential activities of TCMs.

A machine learning model using SNPs obtained from a genome-wide association study predicts the onset of vincristine-induced peripheral neuropathy

Vincristine treatment may cause peripheral neuropathy. In this study, we identified the genes associated with the development of peripheral neuropathy due to vincristine therapy using a genome-wide association study (GWAS) and constructed a predictive model for the development of peripheral neuropathy using genetic information-based machine learning. The study included 72 patients admitted to the Department of Hematology, Tokushima University Hospital, who received vincristine. Of these, 56 were genotyped using the Illumina Asian Screening Array-24 Kit, and a GWAS for the onset of peripheral neuropathy caused by vincristine was conducted. Using Sanger sequencing for 16 validation samples, the top three single nucleotide polymorphisms (SNPs) associated with the onset of peripheral neuropathy were determined. Machine learning was performed using the statistical software R package "caret". The 56 GWAS and 16 validation samples were used as the training and test sets, respectively. Predictive models were constructed using random forest, support vector machine, naive Bayes, and neural network algorithms. According to the GWAS, rs2110179, rs7126100, and rs2076549 were associated with the development of peripheral neuropathy on vincristine administration. Machine learning was performed using these three SNPs to construct a prediction model. A high accuracy of 93.8% was obtained with the support vector machine and neural network using rs2110179 and rs2076549. Thus, peripheral neuropathy development due to vincristine therapy can be effectively predicted by a machine learning prediction model using SNPs associated with it.

Development of binary classification models for assessment of drug-induced liver injury in humans using a large set of FDA-approved drugs

Drug-induced liver injury (DILI) has been identified as one of the major causes for drugs withdrawn from the market, and even termination during the late stages of development. Therefore, it is imperative to evaluate the DILI potential of lead compounds during the research and development process. Although various computational models have been developed to predict DILI, most of which applied the DILI data were extracted from preclinical sources. In this investigation, the in silico prediction models for DILI were constructed based on 1140 FDA-approved drugs by using naïve Bayes classifier approach. The genetic algorithm method was applied for the molecular descriptors selection. Among these established prediction models, the NB-11 model based on eight molecular descriptors combined with ECFP_18 showed the best prediction performance for DILI, which gave 91.7% overall prediction accuracy for the training set, and 68.9% concordance for the external test set. Therefore, the established NB-11 prediction model can be used as a reliable virtual screening tool to predict DILI adverse effect in the early stages of drug design. In addition, some new structural alters for DILI were identified, which could be used for structural optimization in the future drug design by medicinal chemists.

An Algorithm Framework for Drug-Induced Liver Injury Prediction Based on Genetic Algorithm and Ensemble Learning

In the process of drug discovery, drug-induced liver injury (DILI) is still an active research field and is one of the most common and important issues in toxicity evaluation research. It directly leads to the high wear attrition of the drug. At present, there are a variety of computer algorithms based on molecular representations to predict DILI. It is found that a single molecular representation method is insufficient to complete the task of toxicity prediction, and multiple molecular fingerprint fusion methods have been used as model input. In order to solve the problem of high dimensional and unbalanced DILI prediction data, this paper integrates existing datasets and designs a new algorithm framework, Rotation-Ensemble-GA (R-E-GA). The main idea is to find a feature subset with better predictive performance after rotating the fusion vector of high-dimensional molecular representation in the feature space. Then, an Adaboost-type ensemble learning method is integrated into R-E-GA to improve the prediction accuracy. The experimental results show that the performance of R-E-GA is better than other state-of-art algorithms including ensemble learning-based and graph neural network-based methods. Through five-fold cross-validation, the R-E-GA obtains an ACC of 0.77, an F1 score of 0.769, and an AUC of 0.842.

Drug-Induced Immune Thrombocytopenia Toxicity Prediction Based on Machine Learning

Drug-induced immune thrombocytopenia (DITP) often occurs in patients receiving many drug treatments simultaneously. However, clinicians usually fail to accurately distinguish which drugs can be plausible culprits. Despite significant advances in laboratory-based DITP testing, in vitro experimental assays have been expensive and, in certain cases, cannot provide a timely diagnosis to patients. To address these shortcomings, this paper proposes an efficient machine learning-based method for DITP toxicity prediction. A small dataset consisting of 225 molecules was constructed. The molecules were represented by six fingerprints, three descriptors, and their combinations. Seven classical machine learning-based models were examined to determine an optimal model. The results show that the RDMD + PubChem-k-NN model provides the best prediction performance among all the models, achieving an area under the curve of 76.9% and overall accuracy of 75.6% on the external validation set. The application domain (AD) analysis demonstrates the prediction reliability of the RDMD + PubChem-k-NN model. Five structural fragments related to the DITP toxicity are identified through information gain (IG) method along with fragment frequency analysis. Overall, as far as known, it is the first machine learning-based classification model for recognizing chemicals with DITP toxicity and can be used as an efficient tool in drug design and clinical therapy.

In Silico and In Vitro Evaluations of Fluorophoric Thiazolo-[2,3-b]quinazolinones as Anti-cancer Agents Targeting EGFR-TKD

Epidermal growth factor receptor tyrosine kinase domain (EGFR-TKD) plays a pivotal role in cellular signaling, growth, and metabolism. The EGFR-TKD is highly expressed in cancer cells and was endorsed as a therapeutic target for cancer management to overcome metastasis, cell proliferation, and angiogenesis. The novel thiazolo-[2,3-b]quinazolinones series were strategically developed by microwave-assisted organic synthesis and multi dominos reactions aimed to identify the potent thiazolo-[2,3-b]quinazolinone inhibitor against EGFR-TKD. This study explores the binding stability and binding strength of newly developed series via molecular docking, molecular dynamics simulation, and MM/PBSA and MM/GBSA calculations. The binding interaction was observed to be through the functional groups on aryl substituents at positions 3 and 5 of the thiazolo-[2, 3-b]quinazolinone scaffold. The methyl substituents at position 8 of the ligands had prominent hydrophobic interactions corroborating their bindings similar to the reference FDA-approved drug erlotinib in the active site. ADMET predictions reveal that derivatives 5ab, 5aq, and 5bq are drug-like and may be effective in in vitro study. Molecular dynamics simulation for 100 ns of docked complexes revealed their stability at the atomistic level. The ΔGbinding of thiazolo-[2,3-b]quinazolinone was found to be 5ab - 22.45, 5aq - 22.23, and 5bq - 20.76 similar to standard drug, and erlotinib - 24.11 kcal/mol was determined by MM/GBSA method. Furthermore, the anti-proliferative activity of leads of thiazolo-[2,3-b]quinazolinones (n = 3) was studied against breast cancer cell line (MCF-7) and non-small lung carcinoma cell line (H-1299). The highest inhibitions in cell proliferation were shown by 5bq derivatives, and the IC₅₀ was found to be 6.5 ± 0.67 µM against MCF-7 and 14.8 µM against H-1299. The noscapine was also taken as a positive control and showed IC₅₀ at higher concentrations 37 ± 1 against MCF-7 and 46.5 ± 1.2 against H-1299.

Microwave-Assisted Synthesis, Biological Activity Evaluation, Molecular Docking, and ADMET Studies of Some Novel Pyrrolo [2,3-b] Pyrrole Derivatives

Novel pyrrolo [2,3-b] pyrrole derivatives were synthesized and their hypolipidemic activity was assessed in hyperlipidemic rats. The chemical structures of the new derivatives were confirmed through spectral analysis. Compounds 5 and 6 were revealed to be the most effective hypolipidemic agents, with considerable hypocholesterolemic and hypotriglyceridemic effects. They appear to be promising candidates for creating new powerful derivatives with anti-atherosclerotic and hypolipidemic properties. As for antimicrobial activity, some of the tested compounds showed moderate activity against Pseudomonas aeruginosa: compound 2 revealed an MIC value of 50 μg/mL, compared to 25 μg/mL for ciprofloxacin. Compound 3 showed good antimicrobial activity against Staphylococcus aureus, comparable to ciprofloxacin, and roughly half the activity of ampicillin, according to MIC values. Compound 2 has an MIC approximately 25% of that of clotrimazole against Candida albicans. Compound 2 also showed the highest antioxidant activity with 59% inhibition of radical scavenging activity. Additionally, the cytotoxic activity of these new derivatives 1–7 was investigated and most of them showed good anticancer activity against the three tested cell lines.

Development of in silico prediction models for drug-induced liver malignant tumors based on the activity of molecular initiating events: Biologically interpretable features

Liver malignant tumors (LMTs) have recently been reported as severe and life-threatening adverse drug events associated with drug-induced liver injury (DILI). DILIs are the most common adverse drug event and can cause the withdrawal of medicinal products or major regulatory action. To reduce the attrition rate and cost of drug discovery, various quantitative structure-toxicity relationship models have been proposed to predict the probability of a DILI based on the chemical structure of a drug. However, there are many unresolved issues regarding the predictors of LMT-inducing drugs, and biologically interpretable prediction models for LMT have not been developed. Here, we constructed prediction models for whether a drug is LMT-inducing based on the activity of molecular initiating events (MIEs), which are biologically interpretable features and are defined as the initial interaction between a molecule and biosystem. We then constructed five machine learning models (i.e., LightGBM, XGBoost, random forest, neural network, and support vector machine) and evaluated their predictive performances. LightGBM achieved the best performance among the tested models. The MIEs making the highest contribution to the model construction for drug-induced LMT were inducement of Enhanced Level of Genome Instability Gene 1 (human ATAD5), nuclear factor-κ B, and activation of thyrotropin-releasing hormone receptor. These results support the previous literature and can be related to the mechanism onset of drug-induced LMT. Our findings may provide useful knowledge for drug development, research, and regulatory decision-making and will contribute to building more accurate and meaningful DILI prediction models by increasing understanding of biological predictors.

Integrated Machine Learning and Chemoinformatics-Based Screening of Mycotic Compounds against Kinesin Spindle ProteinEg5 for Lung Cancer Therapy

Among the various types of cancer, lung cancer is the second most-diagnosed cancer worldwide. The kinesin spindle protein, Eg5, is a vital protein behind bipolar mitotic spindle establishment and maintenance during mitosis. Eg5 has been reported to contribute to cancer cell migration and angiogenesis impairment and has no role in resting, non-dividing cells. Thus, it could be considered as a vital target against several cancers, such as renal cancer, lung cancer, urothelial carcinoma, prostate cancer, squamous cell carcinoma, etc. In recent years, fungal secondary metabolites from the Indian Himalayan Region (IHR) have been identified as an important lead source in the drug development pipeline. Therefore, the present study aims to identify potential mycotic secondary metabolites against the Eg5 protein by applying integrated machine learning, chemoinformatics based in silico-screening methods and molecular dynamic simulation targeting lung cancer. Initially, a library of 1830 mycotic secondary metabolites was screened by a predictive machine-learning model developed based on the random forest algorithm with high sensitivity (1) and an ROC area of 0.99. Further, 319 out of 1830 compounds screened with active potential by the model were evaluated for their drug-likeness properties by applying four filters simultaneously, viz., Lipinski's rule, CMC-50 like rule, Veber rule, and Ghose filter. A total of 13 compounds passed from all the above filters were considered for molecular docking, functional group analysis, and cell line cytotoxicity prediction. Finally, four hit mycotic secondary metabolites found in fungi from the IHR were screened viz., (-)-Cochlactone-A, Phelligridin C, Sterenin E, and Cyathusal A. All compounds have efficient binding potential with Eg5, containing functional groups like aromatic rings, rings, carboxylic acid esters, and carbonyl and with cell line cytotoxicity against lung cancer cell lines, namely, MCF-7, NCI-H226, NCI-H522, A549, and NCI H187. Further, the molecular dynamics simulation study confirms the docked complex rigidity and stability by exploring root mean square deviations, root mean square fluctuations, and radius of gyration analysis from 100 ns simulation trajectories. The screened compounds could be used further to develop effective drugs against lung and other types of cancer.

Relationships between the Structure and Severe Drug-Induced Liver Injury for Low, Medium, and High Doses of Drugs

Assessment of structure-activity relationships (SARs) for predicting severe drug-induced liver injury (DILI) is essential since in vivo and in vitro preclinical methods cannot detect many druglike compounds disrupting liver functions. To date, plenty of SAR models for the prediction of DILI have been developed; however, none of them considered the route of drug administration and daily dose, which may introduce significant bias into prediction results. We have created a dataset of 617 drugs with parenteral and oral administration routes and consistent information on DILI severity. We have found a clear relationship between route, dose, and DILI severity. According to SAR, nearly 40% of moderate- and non-DILI-causing drugs would cause severe DILI if they were administered at high oral doses. We have proposed the following approach to predict severe DILI. New compounds recommended to be used at low oral doses (<∼10 mg daily), or parenterally, can be considered not causing severe DILI. DILI for compounds administered at medium oral doses (∼10-100 mg daily; 22.2% of drugs under consideration) can be considered unpredictable because reasonable SAR models were not obtained due to the small size and heterogeneity of the corresponding dataset. The DILI potential of the compounds recommended to be used at high oral doses (more than ∼100 mg daily) can be estimated using SAR modeling. The balanced accuracy of the approach calculated by a 10-fold cross-validation procedure is 0.803. The developed approach can be used to estimate severe DILI for druglike compounds proposed to use at low and high oral doses or parenterally at the early stages of drug development.

Machine Learning Models for Predicting Liver Toxicity

Liver toxicity is a major adverse drug reaction that accounts for drug failure in clinical trials and withdrawal from the market. Therefore, predicting potential liver toxicity at an early stage in drug discovery is crucial to reduce costs and the potential for drug failure. However, current in vivo animal toxicity testing is very expensive and time consuming. As an alternative approach, various machine learning models have been developed to predict potential liver toxicity in humans. This chapter reviews current advances in the development and application of machine learning models for prediction of potential liver toxicity in humans and discusses possible improvements to liver toxicity prediction.

In Silico Models for Hepatotoxicity

In this chapter, we review the state of the art of predicting human hepatotoxicity using in silico techniques. There has been significant progress in this area over the past 20 years but there are still some challenges ahead. Principally, these challenges are our partial understanding of a very complex biochemical system and our ability to emulate that in a predictive capacity. Here, we provide an overview of the published modeling approaches in this area to date and discuss their design, strengths and weaknesses. It is interesting to note the diversity in modeling approaches, whether they be statistical algorithms or evidenced-based approaches including structural alerts and pharmacophore models. Irrespective of modeling approach, it appears a common theme of access to appropriate, relevant, and high-quality data is a limitation to all and is likely to continue to be the focus of future research.

Prediction of Drug-Induced Liver Toxicity Using SVM and Optimal Descriptor Sets

Drug-induced liver toxicity is one of the significant safety challenges for the patient's health and the pharmaceutical industry. It causes termination of drug candidates in clinical trials and also the retractions of approved drugs from the market. Thus, it is essential to identify hepatotoxic compounds in the initial stages of drug development process. The purpose of this study is to construct quantitative structure activity relationship models using machine learning algorithms and systematical feature selection methods for molecular descriptor sets. The models were built from a large and diverse set of 1253 drug compounds and were validated internally with 10-fold cross-validation. In this study, we applied a variety of feature selection techniques to extract the optimal subset of descriptors as modeling features to improve the prediction performance. Experimental results suggested that the support vector machine-based classifier had achieved a better classification accuracy with reduced molecular descriptors. The final optimal model provides an accuracy of 0.811, a sensitivity of 0.840, a specificity of 0.783 and Mathew's correlation coefficient of 0.623 with an internal validation set. Furthermore, this model outperformed the prior studies while evaluated in both the internal and external test sets. The utilization of distinct optimal molecular descriptors as modeling features produce an in silico model with a superior performance.

ChemBioSim: Enhancing Conformal Prediction of In Vivo Toxicity by Use of Predicted Bioactivities

Computational methods such as machine learning approaches have a strong track record of success in predicting the outcomes of in vitro assays. In contrast, their ability to predict in vivo endpoints is more limited due to the high number of parameters and processes that may influence the outcome. Recent studies have shown that the combination of chemical and biological data can yield better models for in vivo endpoints. The ChemBioSim approach presented in this work aims to enhance the performance of conformal prediction models for in vivo endpoints by combining chemical information with (predicted) bioactivity assay outcomes. Three in vivo toxicological endpoints, capturing genotoxic (MNT), hepatic (DILI), and cardiological (DICC) issues, were selected for this study due to their high relevance for the registration and authorization of new compounds. Since the sparsity of available biological assay data is challenging for predictive modeling, predicted bioactivity descriptors were introduced instead. Thus, a machine learning model for each of the 373 collected biological assays was trained and applied on the compounds of the in vivo toxicity data sets. Besides the chemical descriptors (molecular fingerprints and physicochemical properties), these predicted bioactivities served as descriptors for the models of the three in vivo endpoints. For this study, a workflow based on a conformal prediction framework (a method for confidence estimation) built on random forest models was developed. Furthermore, the most relevant chemical and bioactivity descriptors for each in vivo endpoint were preselected with lasso models. The incorporation of bioactivity descriptors increased the mean F1 scores of the MNT model from 0.61 to 0.70 and for the DICC model from 0.72 to 0.82 while the mean efficiencies increased by roughly 0.10 for both endpoints. In contrast, for the DILI endpoint, no significant improvement in model performance was observed. Besides pure performance improvements, an analysis of the most important bioactivity features allowed detection of novel and less intuitive relationships between the predicted biological assay outcomes used as descriptors and the in vivo endpoints. This study presents how the prediction of in vivo toxicity endpoints can be improved by the incorporation of biological information-which is not necessarily captured by chemical descriptors-in an automated workflow without the need for adding experimental workload for the generation of bioactivity descriptors as predicted outcomes of bioactivity assays were utilized. All bioactivity CP models for deriving the predicted bioactivities, as well as the in vivo toxicity CP models, can be freely downloaded from https://doi.org/10.5281/zenodo.4761225.

Development and Application of a Transcriptomic Signature of Bioactivation in an Advanced In Vitro Liver Model to Reduce Drug-induced Liver Injury Risk Early in the Pharmaceutical Pipeline

Early risk assessment of drug-induced liver injury (DILI) potential for drug candidates remains a major challenge for pharmaceutical development. We have previously developed a set of rat liver transcriptional biomarkers in short-term toxicity studies to inform the potential of drug candidates to generate a high burden of chemically reactive metabolites that presents higher risk for human DILI. Here, we describe translation of those NRF1-/NRF2-mediated liver tissue biomarkers to an in vitro assay using an advanced micropatterned coculture system (HEPATOPAC) with primary hepatocytes from male Wistar Han rats. A 9-day, resource-sparing and higher throughput approach designed to identify new chemical entities with lower reactive metabolite-forming potential was qualified for internal decision making using 93 DILI-positive and -negative drugs. This assay provides 81% sensitivity and 90% specificity in detecting hepatotoxicants when a positive test outcome is defined as the bioactivation signature score of a test drug exceeding the threshold value at an in vitro test concentration that falls within 3-fold of the estimated maximum drug concentration at the human liver inlet following highest recommended clinical dose administrations. Using paired examples of compounds from distinct chemical series and close structural analogs, we demonstrate that this assay can differentiate drugs with lower DILI risk. The utility of this in vitro transcriptomic approach was also examined using human HEPATOPAC from a single donor, yielding 68% sensitivity and 86% specificity when the aforementioned criteria are applied to the same 93-drug test set. Routine use of the rat model has been adopted with deployment of the human model as warranted on a case-by-case basis. This in vitro transcriptomic signature-based strategy can be used early in drug discovery to derisk DILI potential from chemically reactive metabolites by guiding structure-activity relationship hypotheses and candidate selection.

Modeling approaches for reducing safety-related attrition in drug discovery and development: a review on myelotoxicity, immunotoxicity, cardiovascular toxicity, and liver toxicity

Introduction:Safety and tolerability is a critical area where improvements are needed to decrease the attrition rates during development of new drug candidates. Modeling approaches, when smartly implemented, can contribute to this aim.Areas covered:The focus of this review was on modeling approaches applied to four kinds of drug-induced toxicities: hematological, immunological, cardiovascular (CV) and liver toxicity. Papers, mainly published in the last 10 years, reporting models in three main methodological categories - computational models (e.g., quantitative structure-property relationships, machine learning approaches, neural networks, etc.), pharmacokinetic-pharmacodynamic (PK-PD) models, and quantitative system pharmacology (QSP) models - have been considered.Expert opinion:The picture observed in the four examined toxicity areas appears heterogeneous. Computational models are typically used in all areas as screening tools in the early stages of development for hematological, cardiovascular and liver toxicity, with accuracies in the range of 70-90%. A limited number of computational models, based on the analysis of drug protein sequence, was instead proposed for immunotoxicity. In the later stages of development, toxicities are quantitatively predicted with reasonably good accuracy using either semi-mechanistic PK-PD models (hematological and cardiovascular toxicity), or fully exploited QSP models (immuno-toxicity and liver toxicity).

Determination of "fitness-for-purpose" of quantitative structure-activity relationship (QSAR) models to predict (eco-)toxicological endpoints for regulatory use

In silico models are used to predict toxicity and molecular properties in chemical safety assessment, gaining widespread regulatory use under a number of legislations globally. This study has rationalised previously published criteria to evaluate quantitative structure-activity relationships (QSARs) in terms of their uncertainty, variability and potential areas of bias, into ten assessment components, or higher level groupings. The components have been mapped onto specific regulatory uses (i.e. data gap filling for risk assessment, classification and labelling, and screening and prioritisation) identifying different levels of uncertainty that may be acceptable for each. Twelve published QSARs were evaluated using the components, such that their potential use could be identified. High uncertainty was commonly observed with the presentation of data, mechanistic interpretability, incorporation of toxicokinetics and the relevance of the data for regulatory purposes. The assessment components help to guide strategies that can be implemented to improve acceptability of QSARs through the reduction of uncertainties. It is anticipated that model developers could apply the assessment components from the model design phase (e.g. through problem formulation) through to their documentation and use. The application of the components provides the possibility to assess QSARs in a meaningful manner and demonstrate their fitness-for-purpose against pre-defined criteria.

The Promise of AI for DILI Prediction

Drug-induced liver injury (DILI) is a common reason for the withdrawal of a drug from the market. Early assessment of DILI risk is an essential part of drug development, but it is rendered challenging prior to clinical trials by the complex factors that give rise to liver damage. Artificial intelligence (AI) approaches, particularly those building on machine learning, range from random forests to more recent techniques such as deep learning, and provide tools that can analyze chemical compounds and accurately predict some of their properties based purely on their structure. This article reviews existing AI approaches to predicting DILI and elaborates on the challenges that arise from the as yet limited availability of data. Future directions are discussed focusing on rich data modalities, such as 3D spheroids, and the slow but steady increase in drugs annotated with DILI risk labels.

Rodent models of cholestatic liver disease: A practical guide for translational research

Cholestatic liver disease denotes any situation associated with impaired bile flow concomitant with a noxious bile acid accumulation in the liver and/or systemic circulation. Cholestatic liver disease can be subdivided into different types according to its clinical phenotype, such as biliary atresia, drug-induced cholestasis, gallstone liver disease, intrahepatic cholestasis of pregnancy, primary biliary cholangitis and primary sclerosing cholangitis. Considerable effort has been devoted to elucidating underlying mechanisms of cholestatic liver injuries and explore novel therapeutic and diagnostic strategies using animal models. Animal models employed according to their appropriate applicability domain herein play a crucial role. This review provides an overview of currently available in vivo animal models, fit-for-purpose in modelling different types of cholestatic liver diseases. Moreover, a practical guide and workflow is provided which can be used for translational research purposes, including all advantages and disadvantages of currently available in vivo animal models.

Prediction and mechanistic analysis of drug-induced liver injury (DILI) based on chemical structure

BACKGROUND

Drug-induced liver injury (DILI) is a major safety concern characterized by a complex and diverse pathogenesis. In order to identify DILI early in drug development, a better understanding of the injury and models with better predictivity are urgently needed. One approach in this regard are in silico models which aim at predicting the risk of DILI based on the compound structure. However, these models do not yet show sufficient predictive performance or interpretability to be useful for decision making by themselves, the former partially stemming from the underlying problem of labeling the in vivo DILI risk of compounds in a meaningful way for generating machine learning models.

RESULTS

As part of the Critical Assessment of Massive Data Analysis (CAMDA) "CMap Drug Safety Challenge" 2019 ( http://camda2019.bioinf.jku.at ), chemical structure-based models were generated using the binarized DILIrank annotations. Support Vector Machine (SVM) and Random Forest (RF) classifiers showed comparable performance to previously published models with a mean balanced accuracy over models generated using 5-fold LOCO-CV inside a 10-fold training scheme of 0.759 ± 0.027 when predicting an external test set. In the models which used predicted protein targets as compound descriptors, we identified the most information-rich proteins which agreed with the mechanisms of action and toxicity of nonsteroidal anti-inflammatory drugs (NSAIDs), one of the most important drug classes causing DILI, stress response via TP53 and biotransformation. In addition, we identified multiple proteins involved in xenobiotic metabolism which could be novel DILI-related off-targets, such as CLK1 and DYRK2. Moreover, we derived potential structural alerts for DILI with high precision, including furan and hydrazine derivatives; however, all derived alerts were present in approved drugs and were over specific indicating the need to consider quantitative variables such as dose.

CONCLUSION

Using chemical structure-based descriptors such as structural fingerprints and predicted protein targets, DILI prediction models were built with a predictive performance comparable to previous literature. In addition, we derived insights on proteins and pathways statistically (and potentially causally) linked to DILI from these models and inferred new structural alerts related to this adverse endpoint.

Artificial Intelligence in Drug Discovery: A Comprehensive Review of Data-driven and Machine Learning Approaches

As expenditure on drug development increases exponentially, the overall drug discovery process requires a sustainable revolution. Since artificial intelligence (AI) is leading the fourth industrial revolution, AI can be considered as a viable solution for unstable drug research and development. Generally, AI is applied to fields with sufficient data such as computer vision and natural language processing, but there are many efforts to revolutionize the existing drug discovery process by applying AI. This review provides a comprehensive, organized summary of the recent research trends in AI-guided drug discovery process including target identification, hit identification, ADMET prediction, lead optimization, and drug repositioning. The main data sources in each field are also summarized in this review. In addition, an in-depth analysis of the remaining challenges and limitations will be provided, and proposals for promising future directions in each of the aforementioned areas.

Predicting the reproductive toxicity of chemicals using ensemble learning methods and molecular fingerprints

Reproductive toxicity endpoints are a significant safety concern in the assessment of the adverse effects of chemicals in drug discovery. Computational models that can accurately predict a chemical's toxic potential are increasingly pursued to replace traditional animal experiments. Thus, ensemble learning models were built to predict the reproductive toxicity of compounds. Our ensemble models were developed using support vector machine, random forest, and extreme gradient boosting methods and 9 molecular fingerprints calculated for a dataset containing 1823 chemicals. The best prediction performance was achieved by the Ensemble-Top12 model, with an accuracy (ACC) of 86.33 %, a sensitivity (SEN) of 82.02 %, a specificity (SPE) of 90.19 %, and an area under the receiver operating characteristic curve (AUC) of 0.937 in 5-fold cross-validation and ACC, SEN, SPE, and AUC values of 84.38 %, 86.90 %, 90.67 %, and 0.920, respectively, in external validation. We also defined the applicability domain (AD) of the ensemble model by calculating the Tanimoto distance of the training set. Compared with models in existing literature, our ensemble model achieves relatively high ACC, SPE and AUC values. We also identified several fingerprint features related to chemical reproductive toxicity. Considering the performance of model, we recommend using the Ensemble-Top12 model to predict reproductive toxicity in early drug development.

Quantitative Structure-Activity Relationship (QSAR) Model for the Severity Prediction of Drug-Induced Rhabdomyolysis by Using Random Forest

Drug-induced rhabdomyolysis (DIR) is a rare and potentially life-threatening muscle injury that is characterized by low incidence and high risk. To our best knowledge, the performance of the current predictive models for the early detection of DIR is suboptimal because of the scarcity and dispersion of DIR cases. Therefore, on the basis of the curated drug information from the Drug-Induced Rhabdomyolysis Atlas (DIRA) database, we proposed a random forest (RF) model to predict the DIR severity of the marketed drugs. Compared with the state-of-art methods, our proposed model outperformed extreme gradient boosting, support vector machine, and logistic regression in distinguishing the Most-DIR concern drugs from the No-DIR concern drugs (Matthews correlation coefficient (MCC) and recall rate of our model were 0.46 and 0.81, respectively). Our model was subsequently applied to predicting the potentially serious DIR for 1402 drugs, which were reported to cause DIR by the postmarketing DIR surveillance data in the FDA Spontaneous Adverse Events Reporting System (FAERS). As a result, 62.7% (94) of drugs ranked in the top 150 drugs with the Most-DIR concerns in FAERS can be identified by our model. The top four drugs (odds ratio >30) including acepromazine, rapacuronium, oxyphenbutazone, and naringenin were correctly predicted by our model. In conclusion, the RF model can well predict the Most-DIR concern drug only based on the chemical structure information and can be a facilitated tool for early DIR detection.

Synthesis, molecular docking, and in silico ADME/Tox profiling studies of new 1-aryl-5-(3-azidopropyl)indol-4-ones: Potential inhibitors of SARS CoV-2 main protease

The virus SARS CoV-2, which causes the respiratory infection COVID-19, continues its spread across the world and to date has caused more than a million deaths. Although COVID-19 vaccine development appears to be progressing rapidly, scientists continue the search for different therapeutic options to treat this new illness. In this work, we synthesized five new 1-aryl-5-(3-azidopropyl)indol-4-ones and showed them to be potential inhibitors of the SARS CoV-2 main protease (3CLpro). The compounds were obtained in good overall yields and molecular docking indicated favorable binding with 3CLpro. In silico ADME/Tox profile of the new compounds were calculated using the SwissADME and pkCSM-pharmacokinetics web tools, and indicated adequate values of absorption, distribution and excretion, features related to bioavailability. Moreover, low values of toxicity were indicated for these compounds. And drug-likeness levels of the compounds were also predicted according to the Lipinski and Veber rules.

Machine Learning in Predictive Toxicology: Recent Applications and Future Directions for Classification Models

In recent times, machine learning has become increasingly prominent in predictive toxicology as it has shifted from in vivo studies toward in silico studies. Currently, in vitro methods together with other computational methods such as quantitative structure-activity relationship modeling and absorption, distribution, metabolism, and excretion calculations are being used. An overview of machine learning and its applications in predictive toxicology is presented here, including support vector machines (SVMs), random forest (RF) and decision trees (DTs), neural networks, regression models, naïve Bayes, k-nearest neighbors, and ensemble learning. The recent successes of these machine learning methods in predictive toxicology are summarized, and a comparison of some models used in predictive toxicology is presented. In predictive toxicology, SVMs, RF, and DTs are the dominant machine learning methods due to the characteristics of the data available. Lastly, this review describes the current challenges facing the use of machine learning in predictive toxicology and offers insights into the possible areas of improvement in the field.

DeepDILI: Deep Learning-Powered Drug-Induced Liver Injury Prediction Using Model-Level Representation

Drug-induced liver injury (DILI) is the most frequently reported single cause of safety-related withdrawal of marketed drugs. It is essential to identify drugs with DILI potential at the early stages of drug development. In this study, we describe a deep learning-powered DILI (DeepDILI) prediction model created by combining model-level representation generated by conventional machine learning (ML) algorithms with a deep learning framework based on Mold2 descriptors. We conducted a comprehensive evaluation of the proposed DeepDILI model performance by posing several critical questions: (1) Could the DILI potential of newly approved drugs be predicted by accumulated knowledge of early approved ones? (2) is model-level representation more informative than molecule-based representation for DILI prediction? and (3) could improved model explainability be established? For question 1, we developed the DeepDILI model using drugs approved before 1997 to predict the DILI potential of those approved thereafter. As a result, the DeepDILI model outperformed the five conventional ML algorithms and two state-of-the-art ensemble methods with a Matthews correlation coefficient (MCC) value of 0.331. For question 2, we demonstrated that the DeepDILI model's performance was significantly improved (i.e., a MCC improvement of 25.86% in test set) compared with deep neural networks based on molecule-based representation. For question 3, we found 21 chemical descriptors that were enriched, suggesting a strong association with DILI outcome. Furthermore, we found that the DeepDILI model has more discrimination power to identify the DILI potential of drugs belonging to the World Health Organization therapeutic category of 'alimentary tract and metabolism'. Moreover, the DeepDILI model based on Mold2 descriptors outperformed the ones with Mol2vec and MACCS descriptors. Finally, the DeepDILI model was applied to the recent real-world problem of predicting any DILI concern for potential COVID-19 treatments from repositioning drug candidates. Altogether, this developed DeepDILI model could serve as a promising tool for screening for DILI risk of compounds in the preclinical setting, and the DeepDILI model is publicly available through https://github.com/TingLi2016/DeepDILI.